Fractional and Self-Adaptive Autoregressive Dragonfly Optimization for Privacy Preserved Data Publishing in Mobile Cloud Computing

The advancement in Mobile Cloud Computing (MCC) has gained immense knowledge in computing concept for upcoming generation. The wireless communi-cations enable the integration of cloud computing and mobile to generate MCC. Privacy and security are the major issues faced by MCC while publishing data. This work introduces a technique, named Self-Adaptive Autoregressive Dragonfly Optimization (S-ADO), for addressing the issues by determining the secret key optimally using retrievable data perturbation technique for privacy preserved data publishing in MCC. The retrievable data perturbation is performed using fractional theory and matrix product based model with proposed S-ADO. The proposed S-ADO is developed by modifying ADO by making it self-adaptive. Initially, a fitness function is computed using privacy & utility parameters for determining the optimal differential derivative coefficients. The optimal coefficients are used to generate the secret key by using fractional theory. Then the matrix product based model is adapted to convert original data into privacy preserved data. The secret key derived using utility and privacy functions, is also used to recover the original data. The performance of the proposed S-ADO algorithm shows superior performance with privacy and utility values as 0.7855, and 0.7088 respectively.